\begin{table}[ht]
\centering
\caption{\textbf{Increase in Republican Presidential Vote Share 
After the Mariel Boatlift, Synthetic Control.}\label{tab:synth_yr}}
\begin{tabular}{cccc}
\toprule \toprule
 & \multicolumn{3}{c}{Counties in Control Pool} \\
 & 250 & 555 & Florida \\[1mm]
\midrule
 All Years & 0.07 & 0.05 & 0.03 \\
 & [96.27] & [90.43] & [79.41] \\[2mm]
 1980 & 0.07 & 0.07 & 0.05 \\
 & [96.27] & [95.21] & [91.18] \\[2mm]
 1984 & 0.08 & 0.06 & -0.01 \\
 & [96.27] & [93.09] & [27.94] \\[2mm]
 1988 & 0.09 & 0.07 & 0.00 \\
 & [96.27] & [95.21] & [51.47] \\[2mm]
 1992 & 0.09 & 0.08 & 0.11 \\
 & [97.76] & [97.34] & [94.12] \\[2mm]
 1996 & 0.04 & 0.01 & 0.02 \\
 & [71.64] & [62.23] & [60.29] \\[2mm]
 2000 & 0.08 & 0.00 & 0.03 \\
 & [80.60] & [54.26] & [67.65] \\[2mm]
\bottomrule \bottomrule
\multicolumn{4}{p{.4\textwidth}}{\footnotesize
All estimates calculated holding out counties that border Miami-Dade. 
The 250 or 555 counties most similar to Miami-Dade in terms of 1978 population 
are included in the control pool. 
The percentile of the placebo distribution where Miami-Dade lands is reported in 
square brackets. 
The placebo distribution is found using the procedure discribed in 
Abadie et al (2010).}
\end{tabular}
\end{table}
